We don’t want cost to be a blocker to attend TokioConf.
If you build with Tokio, we want you there.
For every full-price ticket purchased before the end of March, we’ll add another Community Access ticket so someone else in the community can attend.
"I made this" meme, where AI make something, and the human thinks they made it instead.
Humans be like: "I did this".
That is, combining novices with experts tends to make better products than experts can alone.
Just read: "Junior[s] have always been useless."
Hm. I've found Conway (1968) to be true instead:
> We should take particular note of the fact that the quantum jumps in compiler design progress were associated with new groups of people on territory previously trampled chiefly by [experts].
Finally finished my "2025 in review" post. They always take longer than I think.
But it's always so valuable to reflect, follow the story, and extract the lessons.
seanmonstar.com/blog/2025-in...
Early bird pricing for TokioConf 2026 ends in just a few days.
If you’ve been thinking about joining us this April, now’s a great time to grab a ticket before prices increase.
Get your ticket: www.eventbrite.com/e/tokioconf-...
Finally finished my "2025 in review" post. They always take longer than I think.
But it's always so valuable to reflect, follow the story, and extract the lessons.
seanmonstar.com/blog/2025-in...
the copilot thing smells just like google+ a decade ago
Latest #rustlang h2 release v0.4.13 out!
- API support for 1xx responses
- Fixes for DATA frame padding and stream capacity when pending
github.com/hyperium/h2/...
I get questions on reddit and github about why suddenly it's so much harder to build reqwest, and it makes me sad.
I've pointed people at github.com/aws/aws-lc-r...
It looks like it might be the Cargo.lock as an optional feature of some dependencies, cargo can't not include optional weak dependencies currently :(
github.com/rust-lang/ca...
yea i wouldn't actually maintain this. probably easier to fix aws-lc, or fund an even better provider.
i'm just flailing with embarrassment for windows users trying to upgrade reqwest
I'm imagining a native-rustls crate, which uses rustls on unix, and schannel on windows, and the user doesn't have to care, like native-tls.
rustls is definitely much better, and aws-lc-rs is likely a good idea
but i'm frustated at how poor the experience is for windows users
To end the year, a new major version of #rustlang reqwest, v0.13.0 is out! 🚀
- This brings rustls by default, replacing native-tls. 🦀
- Some feature and method cleanup.
- But easy to upgrade for most everyone.
seanmonstar.com/blog/reqwest...
why yes i did want to debug cmake on android
To end the year, a new major version of #rustlang reqwest, v0.13.0 is out! 🚀
- This brings rustls by default, replacing native-tls. 🦀
- Some feature and method cleanup.
- But easy to upgrade for most everyone.
seanmonstar.com/blog/reqwest...
Still appreciate any experience reports!
Assuming nothing is terribly wrong, the hope is to release for real next week.
Just published a new #rustlang reqwest release candidate: v0.13.0-rc.1.
This has some breaking changes, the biggest was switching to rustls by default.
I'd appreciate if you gave it a spin 🙏
github.com/seanmonstar/...
Just published a new #rustlang reqwest release candidate: v0.13.0-rc.1.
This has some breaking changes, the biggest was switching to rustls by default.
I'd appreciate if you gave it a spin 🙏
github.com/seanmonstar/...
New #rustlang reqwest v0.12.27 out now!
- Adds Windows Named Pipe support.
github.com/seanmonstar/...
Just one more day to fill out the hyper user survey for 2025!
There's just one more week left on the @hyper.rs user survey this year!
Announcing @hyper.rs composable pool layers!
We took the various aspects of a connection pool and made them into individual layers to combine as needed.
This is something I've been thinking about for years, I'm so glad it's finally coming out. #rustlang
seanmonstar.com/blog/hyper-u...
Announcing @hyper.rs composable pool layers!
We took the various aspects of a connection pool and made them into individual layers to combine as needed.
This is something I've been thinking about for years, I'm so glad it's finally coming out. #rustlang
seanmonstar.com/blog/hyper-u...
New post: a defense of lock poisoning in Rust.
Followup to recent discussion: decided to write about lock poisoning, looking at the arguments on each side, and informed by our experience at @oxide.computer dealing with the parallel problem of unexpected async cancellations
Please give it a read!
There's 2 more weeks of the hyper user survey for 2025.
Thanks all who have already filled it out! We'll be using this to make things better in 2026 🦀
I'm excited to announce the first @hyper.rs user survey!
If you've used hyper (or related libraries) just a little or a whole bunch, providing feedback is invaluable and should take less than 5 minutes. Give it a go!
seanmonstar.com/blog/hyper-u...
#rustlang
While I never want to _deal_ with poison, I do want the invalid state propagated.
I like one suggestion I've seen:
- lock() just panics on poison, no Result
- try_lock() let's you try to recover from poison for the 1%
Incredibly disappointed (shocked even) that the plan is to make the default Rust mutex not poisonable in the 2027 edition. Poisoning is one of the best examples of Rust focusing on rigor, and removing it from the default mutex would be a massive step backwards.
